Can't load model around 800gb on M2 Max

[bikevit2008]

Hi! Im newby to this using

vitalijsatilov@MacBook-Pro-Vitalij ~ % llama-server -m /Users/vitalijsatilov/Downloads/llama31/meta-llama-405b-inst-bf16-00001-of-00019.gguf      
INFO [                    main] build info | tid="0x1ea3bd8c0" timestamp=1722545885 build=3497 commit="ed9d2854"
INFO [                    main] system info | tid="0x1ea3bd8c0" timestamp=1722545885 n_threads=8 n_threads_batch=-1 total_threads=12 system_info="AVX = 0 | AVX_VNNI = 0 | AVX2 = 0 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | AVX512_BF16 = 0 | FMA = 0 | NEON = 1 | SVE = 0 | ARM_FMA = 1 | F16C = 0 | FP16_VA = 1 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 0 | SSSE3 = 0 | VSX = 0 | MATMUL_INT8 = 0 | LLAMAFILE = 1 | "
llama_model_loader: additional 18 GGUFs metadata loaded.
llama_model_loader: loaded meta data with 31 key-value pairs and 1137 tensors from /Users/vitalijsatilov/Downloads/llama31/meta-llama-405b-inst-bf16-00001-of-00019.gguf (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv   0:                       general.architecture str              = llama
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = Meta-Llama-3.1-405B-Instruct
llama_model_loader: - kv   3:                           general.basename str              = meta
llama_model_loader: - kv   4:                         general.size_label str              = 405B
llama_model_loader: - kv   5:                            general.license str              = llama3.1
llama_model_loader: - kv   6:                               general.tags arr[str,6]       = ["facebook", "meta", "pytorch", "llam...
llama_model_loader: - kv   7:                          general.languages arr[str,8]       = ["en", "de", "fr", "it", "pt", "hi", ...
llama_model_loader: - kv   8:                          llama.block_count u32              = 126
llama_model_loader: - kv   9:                       llama.context_length u32              = 131072
llama_model_loader: - kv  10:                     llama.embedding_length u32              = 16384
llama_model_loader: - kv  11:                  llama.feed_forward_length u32              = 53248
llama_model_loader: - kv  12:                 llama.attention.head_count u32              = 128
llama_model_loader: - kv  13:              llama.attention.head_count_kv u32              = 16
llama_model_loader: - kv  14:                       llama.rope.freq_base f32              = 500000.000000
llama_model_loader: - kv  15:     llama.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  16:                          general.file_type u32              = 32
llama_model_loader: - kv  17:                           llama.vocab_size u32              = 128256
llama_model_loader: - kv  18:                 llama.rope.dimension_count u32              = 128
llama_model_loader: - kv  19:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  20:                         tokenizer.ggml.pre str              = smaug-bpe
llama_model_loader: - kv  21:                      tokenizer.ggml.tokens arr[str,128256]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  22:                  tokenizer.ggml.token_type arr[i32,128256]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  23:                      tokenizer.ggml.merges arr[str,280147]  = ["Ġ Ġ", "Ġ ĠĠĠ", "ĠĠ ĠĠ", "...
llama_model_loader: - kv  24:                tokenizer.ggml.bos_token_id u32              = 128000
llama_model_loader: - kv  25:                tokenizer.ggml.eos_token_id u32              = 128009
llama_model_loader: - kv  26:                    tokenizer.chat_template str              = {% set loop_messages = messages %}{% ...
llama_model_loader: - kv  27:               general.quantization_version u32              = 2
llama_model_loader: - kv  28:                                   split.no u16              = 0
llama_model_loader: - kv  29:                                split.count u16              = 19
llama_model_loader: - kv  30:                        split.tensors.count i32              = 1137
llama_model_loader: - type  f32:  253 tensors
llama_model_loader: - type bf16:  884 tensors
llm_load_vocab: special tokens cache size = 256
llm_load_vocab: token to piece cache size = 0.7999 MB
llm_load_print_meta: format           = GGUF V3 (latest)
llm_load_print_meta: arch             = llama
llm_load_print_meta: vocab type       = BPE
llm_load_print_meta: n_vocab          = 128256
llm_load_print_meta: n_merges         = 280147
llm_load_print_meta: vocab_only       = 0
llm_load_print_meta: n_ctx_train      = 131072
llm_load_print_meta: n_embd           = 16384
llm_load_print_meta: n_layer          = 126
llm_load_print_meta: n_head           = 128
llm_load_print_meta: n_head_kv        = 16
llm_load_print_meta: n_rot            = 128
llm_load_print_meta: n_swa            = 0
llm_load_print_meta: n_embd_head_k    = 128
llm_load_print_meta: n_embd_head_v    = 128
llm_load_print_meta: n_gqa            = 8
llm_load_print_meta: n_embd_k_gqa     = 2048
llm_load_print_meta: n_embd_v_gqa     = 2048
llm_load_print_meta: f_norm_eps       = 0.0e+00
llm_load_print_meta: f_norm_rms_eps   = 1.0e-05
llm_load_print_meta: f_clamp_kqv      = 0.0e+00
llm_load_print_meta: f_max_alibi_bias = 0.0e+00
llm_load_print_meta: f_logit_scale    = 0.0e+00
llm_load_print_meta: n_ff             = 53248
llm_load_print_meta: n_expert         = 0
llm_load_print_meta: n_expert_used    = 0
llm_load_print_meta: causal attn      = 1
llm_load_print_meta: pooling type     = 0
llm_load_print_meta: rope type        = 0
llm_load_print_meta: rope scaling     = linear
llm_load_print_meta: freq_base_train  = 500000.0
llm_load_print_meta: freq_scale_train = 1
llm_load_print_meta: n_ctx_orig_yarn  = 131072
llm_load_print_meta: rope_finetuned   = unknown
llm_load_print_meta: ssm_d_conv       = 0
llm_load_print_meta: ssm_d_inner      = 0
llm_load_print_meta: ssm_d_state      = 0
llm_load_print_meta: ssm_dt_rank      = 0
llm_load_print_meta: model type       = ?B
llm_load_print_meta: model ftype      = BF16
llm_load_print_meta: model params     = 410.08 B
llm_load_print_meta: model size       = 763.84 GiB (16.00 BPW) 
llm_load_print_meta: general.name     = Meta-Llama-3.1-405B-Instruct
llm_load_print_meta: BOS token        = 128000 '<|begin_of_text|>'
llm_load_print_meta: EOS token        = 128009 '<|eot_id|>'
llm_load_print_meta: LF token         = 128 'Ä'
llm_load_print_meta: EOT token        = 128009 '<|eot_id|>'
llm_load_print_meta: max token length = 256
llm_load_tensors: ggml ctx size =    1.06 MiB
ggml_backend_metal_log_allocated_size: allocated buffer, size = 39680.77 MiB, (39680.83 / 73728.00)
ggml_backend_metal_log_allocated_size: allocated buffer, size = 43008.95 MiB, (82689.78 / 73728.00)ggml_backend_metal_log_allocated_size: warning: current allocated size is greater than the recommended max working set size
ggml_backend_metal_log_allocated_size: allocated buffer, size = 43008.83 MiB, (125698.61 / 73728.00)ggml_backend_metal_log_allocated_size: warning: current allocated size is greater than the recommended max working set size
ggml_backend_metal_log_allocated_size: allocated buffer, size = 43008.95 MiB, (168707.56 / 73728.00)ggml_backend_metal_log_allocated_size: warning: current allocated size is greater than the recommended max working set size
ggml_backend_metal_log_allocated_size: allocated buffer, size = 43008.83 MiB, (211716.39 / 73728.00)ggml_backend_metal_log_allocated_size: warning: current allocated size is greater than the recommended max working set size
ggml_backend_metal_log_allocated_size: allocated buffer, size = 43008.95 MiB, (254725.34 / 73728.00)ggml_backend_metal_log_allocated_size: warning: current allocated size is greater than the recommended max working set size
ggml_backend_metal_log_allocated_size: allocated buffer, size = 43008.83 MiB, (297734.17 / 73728.00)ggml_backend_metal_log_allocated_size: warning: current allocated size is greater than the recommended max working set size
ggml_backend_metal_log_allocated_size: allocated buffer, size = 43008.95 MiB, (340743.12 / 73728.00)ggml_backend_metal_log_allocated_size: warning: current allocated size is greater than the recommended max working set size
ggml_backend_metal_buffer_from_ptr: error: failed to allocate buffer, size = 43008.83 MiB
llama_model_load: error loading model: unable to allocate backend metal buffer
llama_load_model_from_file: failed to load model
llama_init_from_gpt_params: error: failed to load model '/Users/vitalijsatilov/Downloads/llama31/meta-llama-405b-inst-bf16-00001-of-00019.gguf'
 ERR [              load_model] unable to load model | tid="0x1ea3bd8c0" timestamp=1722545886 model="/Users/vitalijsatilov/Downloads/llama31/meta-llama-405b-inst-bf16-00001-of-00019.gguf"
vitalijsatilov@MacBook-Pro-Vitalij ~ % 

Can i use params to setup or i really cant goal it?

[BarfingLemurs]

Try llama-server -c 40000 ...

The model is not loaded to ram but reason is the context does load into ram, and is greater than your usable memory from my understanding.